In [1]:
# importing libraries
import pandas as pd 
import numpy as np 
import matplotlib.pyplot as plt
import seaborn as sns
import warnings
warnings.filterwarnings("ignore")
from scipy import stats
from sklearn.preprocessing import StandardScaler
from sklearn.preprocessing import MinMaxScaler
from scipy import stats
from imblearn.combine import SMOTEENN
In [3]:
#reading csv file to dataframe 
df=pd.read_csv("C:\\Users\\muneeba zaheem\\Downloads\\Country-data.csv")
In [4]:
#seeing first five coloumns
df.head()
Out[4]:
country child_mort exports health imports income inflation life_expec total_fer gdpp
0 Afghanistan 90.2 10.0 7.58 44.9 1610 9.44 56.2 5.82 553
1 Albania 16.6 28.0 6.55 48.6 9930 4.49 76.3 1.65 4090
2 Algeria 27.3 38.4 4.17 31.4 12900 16.10 76.5 2.89 4460
3 Angola 119.0 62.3 2.85 42.9 5900 22.40 60.1 6.16 3530
4 Antigua and Barbuda 10.3 45.5 6.03 58.9 19100 1.44 76.8 2.13 12200
In [5]:
#as exports, imports,health are in percentage of gdp so converting them to original form 
df['exports'] = df['exports']*df['gdpp']/100
df['imports'] = df['imports']*df['gdpp']/100
df['health'] = df['health']*df['gdpp']/100
In [6]:
#cheking info of dataframe
df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 167 entries, 0 to 166
Data columns (total 10 columns):
 #   Column      Non-Null Count  Dtype  
---  ------      --------------  -----  
 0   country     167 non-null    object 
 1   child_mort  167 non-null    float64
 2   exports     167 non-null    float64
 3   health      167 non-null    float64
 4   imports     167 non-null    float64
 5   income      167 non-null    int64  
 6   inflation   167 non-null    float64
 7   life_expec  167 non-null    float64
 8   total_fer   167 non-null    float64
 9   gdpp        167 non-null    int64  
dtypes: float64(7), int64(2), object(1)
memory usage: 13.2+ KB
In [7]:
#checking null value count in each column
df.isnull().sum()
Out[7]:
country       0
child_mort    0
exports       0
health        0
imports       0
income        0
inflation     0
life_expec    0
total_fer     0
gdpp          0
dtype: int64
In [8]:
#5 point summery of numerical columns 
df.describe()
Out[8]:
child_mort exports health imports income inflation life_expec total_fer gdpp
count 167.000000 167.000000 167.000000 167.000000 167.000000 167.000000 167.000000 167.000000 167.000000
mean 38.270060 7420.618847 1056.733204 6588.352108 17144.688623 7.781832 70.555689 2.947964 12964.155689
std 40.328931 17973.885795 1801.408906 14710.810418 19278.067698 10.570704 8.893172 1.513848 18328.704809
min 2.600000 1.076920 12.821200 0.651092 609.000000 -4.210000 32.100000 1.150000 231.000000
25% 8.250000 447.140000 78.535500 640.215000 3355.000000 1.810000 65.300000 1.795000 1330.000000
50% 19.300000 1777.440000 321.886000 2045.580000 9960.000000 5.390000 73.100000 2.410000 4660.000000
75% 62.100000 7278.000000 976.940000 7719.600000 22800.000000 10.750000 76.800000 3.880000 14050.000000
max 208.000000 183750.000000 8663.600000 149100.000000 125000.000000 104.000000 82.800000 7.490000 105000.000000
In [9]:
#checking no.of unique counties in country column
df['country'].nunique()
Out[9]:
167
In [10]:
#encoding
from sklearn.preprocessing import LabelEncoder

le = LabelEncoder()

y = le.fit_transform(df['country'] )
In [11]:
#coping the dataframe into another variable for further use
df_copy=df.copy()
In [12]:
#dropping the country column as it was object data type and 167 unique countries
df.drop('country',1,inplace=True)

visualization¶

In [13]:
#boxplot and distplot for checking outliers and distribution of data in each column
j=1
plt.figure(figsize=(20,40))
for i in df.columns:
    plt.subplot(16,2,j)
    sns.boxplot(df[i])
    plt.title(i,color='white')
    j=j+1
    plt.subplot(16,2,j)
    sns.distplot(df[i])
    j=j+1
In [14]:
#along with boxplot plotting swarmplot to know no.of outlier points 
j=1
plt.figure(figsize=(20,40))
for i in df.columns:
    plt.subplot(16,2,j)
    sns.boxplot(df[i])
    plt.title(i,color='white')
    j=j+1
    plt.subplot(16,2,j)
    sns.swarmplot(df[i])
    j=j+1

As there are only few points are present in each column as outliers, so it is better to replace them with mean or median or mode¶

In [15]:
df.skew()
Out[15]:
child_mort    1.450774
exports       6.720171
health        2.526029
imports       6.618500
income        2.231480
inflation     5.154049
life_expec   -0.970996
total_fer     0.967092
gdpp          2.218051
dtype: float64
In [16]:
#removing outliers by replacing them eith mean 
for i in df.columns:
    q1 , q3 =df[i].quantile([0.25,0.75]) 
    iqr = q3-q1
    lf , uf = q1-1.5*iqr , q3+1.5*iqr
    df[i] = np.where(df[i] <lf ,df[i].mean(),df[i])
    df[i] = np.where(df[i] >uf ,df[i].mean(),df[i])
In [17]:
df.skew()
Out[17]:
child_mort    1.167945
exports       1.139095
health        1.103326
imports       1.391470
income        1.038943
inflation     0.894936
life_expec   -0.547204
total_fer     0.928968
gdpp          1.242576
dtype: float64
In [18]:
#checking whether the ouliers are teated or not?
j=1
plt.figure(figsize=(20,40))
for i in df.columns:
    plt.subplot(16,2,j)
    sns.boxplot(df[i])
    plt.title(i,color='white')
    j=j+1
    plt.subplot(16,2,j)
    sns.swarmplot(df[i])
    j=j+1
In [19]:
# importing necessary libraries for model building
from sklearn.decomposition import PCA

#To perform hierarchical clustering
from scipy.cluster.hierarchy import linkage
from scipy.cluster.hierarchy import dendrogram
from scipy.cluster.hierarchy import cut_tree
from sklearn.cluster import KMeans
from sklearn.metrics import silhouette_samples,silhouette_score
In [20]:
# standard scaling
from sklearn.preprocessing import StandardScaler

scaler= StandardScaler()

df = scaler.fit_transform(df)
In [21]:
# finding no.of clusters to take as per inertia
cs = []
for i in range(2,15):
    kmeans = KMeans(n_clusters = i, init = 'k-means++', max_iter = 600, n_init = 10)
    
    kmeans.fit(df)
    
    cs.append(kmeans.inertia_)

# plot the 
plt.plot(range(2,15), cs)
plt.title('The Elbow Method')
plt.xlabel('Number of clusters')
plt.ylabel('Inertia')
plt.show()
In [22]:
# finding no.of principal components with respect to the varience
from sklearn.decomposition import PCA
pca = PCA()
principalComponents = pca.fit_transform(df)
plt.figure()
plt.plot(np.cumsum(pca.explained_variance_ratio_))
plt.xlabel('Number of Components')
plt.ylabel('Variance (%)') #for each component
plt.title('Explained Variance')
plt.show()
In [23]:
# applying PCA 
pca = PCA(n_components=3)
new_data = pca.fit_transform(df)
# This will be the new data fed to the algorithm.
principal_Df = pd.DataFrame(data = new_data
             , columns = ['principal component 1','principal component 2','principal component 3'])
In [24]:
principal_Df.head()
Out[24]:
principal component 1 principal component 2 principal component 3
0 -3.531169 1.338167 -0.315297
1 -0.106481 -1.614102 -0.209965
2 -0.779823 -0.790575 1.601960
3 -3.490239 2.389875 1.763175
4 1.818064 -0.239527 -0.519753
In [25]:
plt.scatter(principal_Df['principal component 1'],principal_Df['principal component 2'])
Out[25]:
<matplotlib.collections.PathCollection at 0x17c696ce710>
In [26]:
# finding silhouette score for different clusters values
for i in range(2,10):
    k_means = KMeans(n_clusters=i, max_iter = 300, init = 'k-means++',random_state = 0,n_init = 10)
    cluster_labels=k_means.fit_predict(principal_Df)
    print(silhouette_score(principal_Df, cluster_labels))

0.4915156341368339
0.4523045687796382
0.3641800740755203
0.3695124251485627
0.33305771186163974
0.35268742984808843
0.3559790764532396
0.365338913139159
In [27]:
# visualizing the silhouette score scatter plots of cluster data
X=principal_Df
import matplotlib.cm as cm 

range_n_clusters = [2,3,4,6,8,10, 14,17, 19,22,24,26,27]
for n_clusters in range_n_clusters:
    # Create a subplot with 1 row and 2 columns
    fig, (ax1, ax2) = plt.subplots(1, 2)
    fig.set_size_inches(18, 7)

    # The 1st subplot is the silhouette plot
    # The silhouette coefficient can range from -1, 1 but in this example all
    # lie within [-0.1, 1]
    ax1.set_xlim([-0.1, 1])
    # The (n_clusters+1)*10 is for inserting blank space between silhouette
    # plots of individual clusters, to demarcate them clearly.
    ax1.set_ylim([0, len(X) + (n_clusters + 1) * 10])

    # Initialize the clusterer with n_clusters value and a random generator
    # seed of 10 for reproducibility.
    clusterer = KMeans(n_clusters=n_clusters, random_state=10)
    cluster_labels = clusterer.fit_predict(X)

    # The silhouette_score gives the average value for all the samples.
    # This gives a perspective into the density and separation of the formed
    # clusters
    silhouette_avg = silhouette_score(X, cluster_labels)
    print("For n_clusters =", n_clusters,
          "The average silhouette_score is :", silhouette_avg)

    # Compute the silhouette scores for each sample
    sample_silhouette_values = silhouette_samples(X, cluster_labels)

    y_lower = 10
    for i in range(n_clusters):
        # Aggregate the silhouette scores for samples belonging to
        # cluster i, and sort them
        ith_cluster_silhouette_values = \
            sample_silhouette_values[cluster_labels == i]

        ith_cluster_silhouette_values.sort()

        size_cluster_i = ith_cluster_silhouette_values.shape[0]
        y_upper = y_lower + size_cluster_i

        color = cm.nipy_spectral(float(i) / n_clusters)
        ax1.fill_betweenx(np.arange(y_lower, y_upper),
                          0, ith_cluster_silhouette_values,
                          facecolor=color, edgecolor=color, alpha=0.7)

        # Label the silhouette plots with their cluster numbers at the middle
        ax1.text(-0.05, y_lower + 0.5 * size_cluster_i, str(i))

        # Compute the new y_lower for next plot
        y_lower = y_upper + 10  # 10 for the 0 samples

    ax1.set_title("The silhouette plot for the various clusters.")
    ax1.set_xlabel("The silhouette coefficient values")
    ax1.set_ylabel("Cluster label")

    # The vertical line for average silhouette score of all the values
    ax1.axvline(x=silhouette_avg, color="red", linestyle="--")

    ax1.set_yticks([])  # Clear the yaxis labels / ticks
    ax1.set_xticks([-0.1, 0, 0.2, 0.4, 0.6, 0.8, 1])

    # 2nd Plot showing the actual clusters formed
    colors = cm.nipy_spectral(cluster_labels.astype(float) / n_clusters)
    ax2.scatter(X['principal component 1'],X['principal component 2'], marker='.', s=30, lw=0, alpha=0.7,
                c=colors, edgecolor='k')

    # Labeling the clusters
    centers = clusterer.cluster_centers_
    # Draw white circles at cluster centers
    ax2.scatter(centers[:,0], centers[:, 1], marker='o',
                c="white", alpha=1, s=200, edgecolor='k')

    for i, c in enumerate(centers):
        ax2.scatter(c[0], c[1], marker='$%d$' % i, alpha=1,
                    s=50, edgecolor='k')

    ax2.set_title("The visualization of the clustered data.")
    ax2.set_xlabel("Feature space for the 1st feature")
    ax2.set_ylabel("Feature space for the 2nd feature")

    plt.suptitle(("Silhouette analysis for KMeans clustering on sample data "
                  "with n_clusters = %d" % n_clusters),
                 fontsize=14, fontweight='bold')

plt.show()

For n_clusters = 2 The average silhouette_score is : 0.4915156341368339
For n_clusters = 3 The average silhouette_score is : 0.4523045687796382
For n_clusters = 4 The average silhouette_score is : 0.3661167171483696
For n_clusters = 6 The average silhouette_score is : 0.32547533409864227
For n_clusters = 8 The average silhouette_score is : 0.36039184694500537
For n_clusters = 10 The average silhouette_score is : 0.3423113217517924
For n_clusters = 14 The average silhouette_score is : 0.32165240875961576
For n_clusters = 17 The average silhouette_score is : 0.33210968170749217
For n_clusters = 19 The average silhouette_score is : 0.3405701437336835
For n_clusters = 22 The average silhouette_score is : 0.33869415898826244
For n_clusters = 24 The average silhouette_score is : 0.34699834971444643
For n_clusters = 26 The average silhouette_score is : 0.3207526846624629
For n_clusters = 27 The average silhouette_score is : 0.3416503076271536

Model building--K_MEANS¶

As we can see that clusters 2,3,4 gives better score and clustering but it's better to take 3 or 4 as no.of clusters¶
In [29]:
# for no.of clusters = 3
k_means = KMeans(n_clusters=3) 

k_means.fit(principal_Df)
Out[29]:
KMeans(n_clusters=3)
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
KMeans(n_clusters=3)
In [30]:
# model parameter study
k_means.cluster_centers_
Out[30]:
array([[-0.51311945, -0.86518452,  0.13565785],
       [ 3.13950623,  0.48738558, -0.01365217],
       [-2.93743544,  0.85928455, -0.21063843]])
In [31]:
# calculate model inertia

k_means.inertia_
Out[31]:
344.4151750150596
In [32]:
k_means.labels_
Out[32]:
array([2, 0, 0, 2, 1, 0, 0, 1, 1, 0, 1, 1, 0, 1, 0, 1, 0, 2, 0, 0, 0, 0,
       0, 1, 0, 2, 2, 0, 2, 1, 0, 2, 2, 1, 0, 0, 2, 2, 2, 0, 2, 1, 1, 1,
       1, 0, 0, 0, 0, 1, 2, 1, 0, 1, 1, 2, 2, 0, 1, 2, 1, 0, 0, 2, 2, 0,
       0, 1, 1, 0, 0, 0, 0, 1, 1, 1, 0, 1, 0, 0, 2, 2, 1, 0, 2, 1, 0, 2,
       2, 1, 1, 1, 0, 2, 2, 1, 0, 2, 1, 2, 0, 0, 0, 0, 0, 0, 2, 0, 2, 0,
       1, 1, 2, 2, 1, 1, 2, 0, 0, 0, 0, 1, 1, 1, 0, 0, 2, 0, 1, 2, 0, 1,
       2, 1, 1, 1, 2, 0, 1, 1, 0, 0, 2, 0, 1, 1, 0, 2, 0, 2, 2, 0, 0, 0,
       0, 2, 0, 1, 1, 1, 0, 0, 0, 0, 0, 2, 2])
In [35]:
# 3D VISUALIZATION
import plotly.express as px
k_means = KMeans(n_clusters=3, max_iter = 300, init = 'k-means++',random_state = 0,n_init = 10)
fig = px.scatter_3d(principal_Df, x='principal component 1', y='principal component 2', z=y, color=k_means.fit_predict(principal_Df), opacity=0.75)
fig.update_layout(coloraxis_colorbar=dict(yanchor="top", y=1, x=0, ticks="outside"))
In [36]:
# for no.of clusters=4
k_means4 = KMeans(n_clusters=4, max_iter = 600, n_init = 10, random_state=7)

k_means4.fit(principal_Df)
Out[36]:
KMeans(max_iter=600, n_clusters=4, random_state=7)
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
KMeans(max_iter=600, n_clusters=4, random_state=7)
In [37]:
k_means4.inertia_
Out[37]:
288.6187520114581
In [38]:
# 3D visualization
k_means = KMeans(n_clusters=4, max_iter = 300, init = 'k-means++',random_state = 0,n_init = 10)
fig = px.scatter_3d(principal_Df, x='principal component 1', y='principal component 2', z=y, color=k_means.fit_predict(principal_Df), opacity=0.75)
fig.update_layout(coloraxis_colorbar=dict(yanchor="top", y=1, x=0, ticks="outside"))

Another way of doing cluster without scaling¶

In [39]:
df_copy2=df_copy.copy()
In [40]:
df_copy.drop('country',axis=1,inplace=True)
In [41]:
# #standardising
# from sklearn.preprocessing import StandardScaler

# scaler= StandardScaler()

# df_co = scaler.fit_transform(df_copy)
# df_copy=pd.DataFrame(df_co,columns=df_copy.columns)
In [42]:
# treating outlies
for i in df_copy.columns:
    q1 , q3 =df_copy[i].quantile([0.25,0.75]) 
    iqr = q3-q1
    lf , uf = q1-1.5*iqr , q3+1.5*iqr
    df_copy[i] = np.where(df_copy[i] <lf ,lf,df_copy[i])
    df_copy[i] = np.where(df_copy[i] >uf ,uf,df_copy[i])
In [43]:
#plotting no.of clusters vs inertia
cs = []
for i in range(2,15):
    kmeans = KMeans(n_clusters = i, init = 'k-means++', max_iter = 600, n_init = 10)
    
    kmeans.fit(df_copy)
    
    cs.append(kmeans.inertia_)

# plot the 
plt.plot(range(2,15), cs)
plt.title('The Elbow Method')
plt.xlabel('Number of clusters')
plt.ylabel('Inertia')
plt.show()
In [44]:
#finding no.of principal components required as per variance
pca = PCA()
principalComponents = pca.fit_transform(df_copy)
plt.figure()
plt.plot(np.cumsum(pca.explained_variance_ratio_))
plt.xlabel('Number of Components')
plt.ylabel('Variance (%)') #for each component
plt.title('Explained Variance')
plt.show()
In [45]:
# Applying PCA
pca = PCA(n_components=2)
new_data = pca.fit_transform(df_copy)
# This will be the new data fed to the algorithm.
principal_Df_copy = pd.DataFrame(data = new_data
             , columns = ['principal component 1','principal component 2'])
In [46]:
#finding silhouette score for each cluster
for i in range(2,10):
    k_means = KMeans(n_clusters=i, max_iter = 300, init = 'k-means++',random_state = 0,n_init = 10)
    cluster_labels=k_means.fit_predict(principal_Df_copy)
    print(silhouette_score(principal_Df_copy, cluster_labels))
#     plt.figure(figsize=(8,6),dpi=100)
#     print(sns.scatterplot(data=principal_Df_copy, x='principal component 1', y='principal component 2', 
#                           hue=cluster_labels, palette='viridis'))

0.7383742667195675
0.6143992722200845
0.5936450171498793
0.5410421329339028
0.5231422923843618
0.5460890141959531
0.5449195967516671
0.539296218512372
In [47]:
# visualizing the silhouette score scatter plots of cluster data
X=principal_Df_copy
import matplotlib.cm as cm 

range_n_clusters = [2,3,4, 10, 14,17, 19,22,24,26,27]
for n_clusters in range_n_clusters:
    # Create a subplot with 1 row and 2 columns
    fig, (ax1, ax2) = plt.subplots(1, 2)
    fig.set_size_inches(18, 7)

    # The 1st subplot is the silhouette plot
    # The silhouette coefficient can range from -1, 1 but in this example all
    # lie within [-0.1, 1]
    ax1.set_xlim([-0.1, 1])
    # The (n_clusters+1)*10 is for inserting blank space between silhouette
    # plots of individual clusters, to demarcate them clearly.
    ax1.set_ylim([0, len(X) + (n_clusters + 1) * 10])

    # Initialize the clusterer with n_clusters value and a random generator
    # seed of 10 for reproducibility.
    clusterer = KMeans(n_clusters=n_clusters, random_state=10)
    cluster_labels = clusterer.fit_predict(X)

    # The silhouette_score gives the average value for all the samples.
    # This gives a perspective into the density and separation of the formed
    # clusters
    silhouette_avg = silhouette_score(X, cluster_labels)
    print("For n_clusters =", n_clusters,
          "The average silhouette_score is :", silhouette_avg)

    # Compute the silhouette scores for each sample
    sample_silhouette_values = silhouette_samples(X, cluster_labels)

    y_lower = 10
    for i in range(n_clusters):
        # Aggregate the silhouette scores for samples belonging to
        # cluster i, and sort them
        ith_cluster_silhouette_values = \
            sample_silhouette_values[cluster_labels == i]

        ith_cluster_silhouette_values.sort()

        size_cluster_i = ith_cluster_silhouette_values.shape[0]
        y_upper = y_lower + size_cluster_i

        color = cm.nipy_spectral(float(i) / n_clusters)
        ax1.fill_betweenx(np.arange(y_lower, y_upper),
                          0, ith_cluster_silhouette_values,
                          facecolor=color, edgecolor=color, alpha=0.7)

        # Label the silhouette plots with their cluster numbers at the middle
        ax1.text(-0.05, y_lower + 0.5 * size_cluster_i, str(i))

        # Compute the new y_lower for next plot
        y_lower = y_upper + 10  # 10 for the 0 samples

    ax1.set_title("The silhouette plot for the various clusters.")
    ax1.set_xlabel("The silhouette coefficient values")
    ax1.set_ylabel("Cluster label")

    # The vertical line for average silhouette score of all the values
    ax1.axvline(x=silhouette_avg, color="red", linestyle="--")

    ax1.set_yticks([])  # Clear the yaxis labels / ticks
    ax1.set_xticks([-0.1, 0, 0.2, 0.4, 0.6, 0.8, 1])

    # 2nd Plot showing the actual clusters formed
    colors = cm.nipy_spectral(cluster_labels.astype(float) / n_clusters)
    ax2.scatter(X['principal component 1'],X['principal component 2'], marker='.', s=30, lw=0, alpha=0.7,
                c=colors, edgecolor='k')

    # Labeling the clusters
    centers = clusterer.cluster_centers_
    # Draw white circles at cluster centers
    ax2.scatter(centers[:,0], centers[:, 1], marker='o',
                c="white", alpha=1, s=200, edgecolor='k')

    for i, c in enumerate(centers):
        ax2.scatter(c[0], c[1], marker='$%d$' % i, alpha=1,
                    s=50, edgecolor='k')

    ax2.set_title("The visualization of the clustered data.")
    ax2.set_xlabel("Feature space for the 1st feature")
    ax2.set_ylabel("Feature space for the 2nd feature")

    plt.suptitle(("Silhouette analysis for KMeans clustering on sample data "
                  "with n_clusters = %d" % n_clusters),
                 fontsize=14, fontweight='bold')

plt.show()

For n_clusters = 2 The average silhouette_score is : 0.7383742667195675
For n_clusters = 3 The average silhouette_score is : 0.6146351449884063
For n_clusters = 4 The average silhouette_score is : 0.5936450171498793
For n_clusters = 10 The average silhouette_score is : 0.5230259898989391
For n_clusters = 14 The average silhouette_score is : 0.4713061496874219
For n_clusters = 17 The average silhouette_score is : 0.4534968607741327
For n_clusters = 19 The average silhouette_score is : 0.4412725261655668
For n_clusters = 22 The average silhouette_score is : 0.43595156069767044
For n_clusters = 24 The average silhouette_score is : 0.4572227121168201
For n_clusters = 26 The average silhouette_score is : 0.42420992580786065
For n_clusters = 27 The average silhouette_score is : 0.4325897866330326

As we can see that among all no.of clusters =3 will gives better results and clsutering and also score is improved to 61.5%¶

In [48]:
# for no.of clusters=3
k_means_copy = KMeans(n_clusters=3, max_iter = 600, n_init = 10, random_state=7)

k_means_copy.fit(principal_Df_copy)
Out[48]:
KMeans(max_iter=600, n_clusters=3, random_state=7)
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
KMeans(max_iter=600, n_clusters=3, random_state=7)
In [49]:
# 3D visualization
import plotly.express as px
k_means = KMeans(n_clusters=3, max_iter = 300, init = 'k-means++',random_state = 0,n_init = 10)
fig = px.scatter_3d(principal_Df_copy, x='principal component 1', y='principal component 2', z=y, color=k_means.fit_predict(principal_Df_copy), opacity=0.75)
fig.update_layout(coloraxis_colorbar=dict(yanchor="top", y=1, x=0, ticks="outside"))

As per the results the clustering without scaling performs well with accuracy of 61%¶

In [50]:
# concatinating the clustering labels with dataframe
country_df_clustered = df_copy2.iloc[:,:]
country_df_clustered = pd.concat([df_copy2, pd.DataFrame(k_means_copy.labels_, columns=['cluster_id_km'])], axis = 1)
# country_df_clustered = pd.concat([df_copy2, pd.DataFrame(k_means_copy_labels, columns=['cluster_id_hc'])], axis = 1)
country_df_clustered.head()
Out[50]:
country child_mort exports health imports income inflation life_expec total_fer gdpp cluster_id_km
0 Afghanistan 90.2 55.30 41.9174 248.297 1610 9.44 56.2 5.82 553 1
1 Albania 16.6 1145.20 267.8950 1987.740 9930 4.49 76.3 1.65 4090 1
2 Algeria 27.3 1712.64 185.9820 1400.440 12900 16.10 76.5 2.89 4460 1
3 Angola 119.0 2199.19 100.6050 1514.370 5900 22.40 60.1 6.16 3530 1
4 Antigua and Barbuda 10.3 5551.00 735.6600 7185.800 19100 1.44 76.8 2.13 12200 2
In [51]:
# printing country names based on there cluster 
print("Cluster 0 of K_means Clustering model : ")
print(country_df_clustered[country_df_clustered['cluster_id_km'] == 0].country.unique())
print('\n')

print("Cluster 1 of K_means Clustering model : ")
print(country_df_clustered[country_df_clustered['cluster_id_km'] == 1].country.unique())
print('\n')


print("Cluster 2 of K_means Clustering model : ")
print(country_df_clustered[country_df_clustered['cluster_id_km'] == 2].country.unique())
print('\n')

Cluster 0 of K_means Clustering model : 
['Australia' 'Austria' 'Bahamas' 'Bahrain' 'Belgium' 'Brunei' 'Canada'
 'Cyprus' 'Czech Republic' 'Denmark' 'Equatorial Guinea' 'Finland'
 'France' 'Germany' 'Greece' 'Iceland' 'Ireland' 'Israel' 'Italy' 'Japan'
 'Kuwait' 'Luxembourg' 'Malta' 'Netherlands' 'New Zealand' 'Norway' 'Oman'
 'Qatar' 'Saudi Arabia' 'Singapore' 'Slovenia' 'South Korea' 'Spain'
 'Sweden' 'Switzerland' 'United Arab Emirates' 'United Kingdom'
 'United States']


Cluster 1 of K_means Clustering model : 
['Afghanistan' 'Albania' 'Algeria' 'Angola' 'Armenia' 'Bangladesh'
 'Belize' 'Benin' 'Bhutan' 'Bolivia' 'Bosnia and Herzegovina'
 'Burkina Faso' 'Burundi' 'Cambodia' 'Cameroon' 'Cape Verde'
 'Central African Republic' 'Chad' 'China' 'Colombia' 'Comoros'
 'Congo, Dem. Rep.' 'Congo, Rep.' "Cote d'Ivoire" 'Dominican Republic'
 'Ecuador' 'Egypt' 'El Salvador' 'Eritrea' 'Fiji' 'Gambia' 'Georgia'
 'Ghana' 'Guatemala' 'Guinea' 'Guinea-Bissau' 'Guyana' 'Haiti' 'India'
 'Indonesia' 'Iraq' 'Jamaica' 'Jordan' 'Kenya' 'Kiribati'
 'Kyrgyz Republic' 'Lao' 'Lesotho' 'Liberia' 'Macedonia, FYR' 'Madagascar'
 'Malawi' 'Mali' 'Mauritania' 'Micronesia, Fed. Sts.' 'Moldova' 'Mongolia'
 'Morocco' 'Mozambique' 'Myanmar' 'Namibia' 'Nepal' 'Niger' 'Nigeria'
 'Pakistan' 'Paraguay' 'Peru' 'Philippines' 'Rwanda' 'Samoa' 'Senegal'
 'Sierra Leone' 'Solomon Islands' 'Sri Lanka'
 'St. Vincent and the Grenadines' 'Sudan' 'Tajikistan' 'Tanzania'
 'Timor-Leste' 'Togo' 'Tonga' 'Tunisia' 'Turkmenistan' 'Uganda' 'Ukraine'
 'Uzbekistan' 'Vanuatu' 'Vietnam' 'Yemen' 'Zambia']


Cluster 2 of K_means Clustering model : 
['Antigua and Barbuda' 'Argentina' 'Azerbaijan' 'Barbados' 'Belarus'
 'Botswana' 'Brazil' 'Bulgaria' 'Chile' 'Costa Rica' 'Croatia' 'Estonia'
 'Gabon' 'Grenada' 'Hungary' 'Iran' 'Kazakhstan' 'Latvia' 'Lebanon'
 'Libya' 'Lithuania' 'Malaysia' 'Maldives' 'Mauritius' 'Montenegro'
 'Panama' 'Poland' 'Portugal' 'Romania' 'Russia' 'Serbia' 'Seychelles'
 'Slovak Republic' 'South Africa' 'Suriname' 'Thailand' 'Turkey' 'Uruguay'
 'Venezuela']
In [ ]: